Home + About me + E-mail me

SI

Why SI?

SI or not SI is one of those arguments that does not have an obvious winner; it all depends on the relative importance you place on different views. Let's look at five points:

• Tradition and history
• Daily life
• Engineering
• International trade
• Difficult calculations and understanding.

Tradition and History

Many of our "old" units can trace their ancestry back thousands of years. The best way of preserving this history might be to continue using them. This argument loses some of its force as many of the old units are already rarely used (perches, quarters, gills) and the rest are gradually being rationalised. In my opinion, the best place to preserve old things is in a museum, but I know others disagree.

In the UK, it does not help that metric is seen as a French invention.

Daily life

There's no doubt that the traditional units are very well designed for day-to-day simple tasks. They are just the right size for measuring drinks, corn, fields, buildings, roads and so on because that is what they were made for. Even the names, typically easy-to-say short words, are more convenient. Anyone whose life does not extend beyond buying apples in the local shop will see no advantage in SI, and that's difficult to gainsay.

Engineering

It is possible to design all sorts of things in either set of units, though there is a strong tendency to rationalise the old units. Sometimes, the SI prefix idea is retro-fitted back onto the old system, to produce micro-inches or kilo-feet. This is a strong sign that the old system is inadequate in many situations where new fields take us away from the human scale. There is also the need to invent new "old" units, such as poundals, slugs and horse-power, to cover measurements that the old system was never designed for.

International trade

Working in the UK, it is now generally possible to assume the world is metric, provided one avoids US suppliers. I have on occasions had to write "warning - inches" on US drawings that we've received. If US stuff comes into our lab with non-metric threads, we have to rummage in the odds-and-ends box to find something to fit. Any company that refuses to recognise metric these days is handicapped, unless it is so big that it controls its marketplace.

Difficult calculations and understanding

Several things attract me to SI, but most important are the clarity of definition, and the internal self-consistency.

The old units were related to human features, or arbitrary, in general could not be defined except by a prototype, and could not be reproduced except by direct reference to the prototype. From the invention of metric, it was seen to be important to relate units to some external constant. The metre was defined as one ten-millionth of the distance from the North pole to the equator through Paris, something not expected to change, and therefore a standard metre could be re-created anywhere at any time. The same was not then possible for the kilogram, which was defined by a prototype, though it could be easily approximated from the metre and the density of water. This process has continued, so that the current definitions [1] are based on atomic properties that are thought to be constant throughout the universe. This is a beautiful idea. We could define our system of units to a green man on Mars, and he could construct examples, without ever seeing the originals.

The self-consistency is even more important. In the old system, new units were invented as they were needed, independently in many fields. Even with simple measurement of volume there were linear-based units such as cubic yards or acre-feet, and special units such as gallons or hogsheads. Often these were originally unrelated, and were only fixed relative to each other fairly recently. Worse happened in the mechanical units, where different units of work (foot-pound, foot-poundal, BTU) and power (horse-power, BTU/hour) arose. Not only are these units unrelated to each other, but they ignore the underlying theory. Take the foot-pound: work is a force times a distance; pound is a unit of mass; the unit assumes that a mass has a fixed weight whereas in reality it varies with location. So either we have to to define the foot-pound at a particular place and time, or measurement of horse-power will vary depending on location. This is horrible.

This lack of relation to the theory also complicates the formulae used to calculate mechanical quantities. Not only are there un-needed conversion constants, but unwanted physical factors, such as the strength of gravity in the power units, crop up too. It is almost always easier, for any non-trivial calculation, to convert everything into SI, calculate, then (if necessary!) convert the answer back.

Old units hinder general understanding. For example, it is customary (in the UK) to measure the mass of babies in pounds, of adults in stones, and of cars in tons or hundredweights; this adds an extra layer which has to removed to see how babies compare with adults, or how much of the proportion of the weight of a car comes from the occupants. Similar problems, but much more significant, occur as the comparisons get more remote from experience.

SI

So, I'm convinced by SI, and have been convinced by metric in general for the last thirty years. Many, though, are influenced by the history argument, and many more by the convenience of old units in simple daily activity. The more one uses measurements, and thinks quantitatively about the world, and talks to people in other walks of life, the more SI is attractive. It will win!

Conversions of old units into SI

(Values in bold are exact.)

List of material properties

References:

1. The Shorter Oxford English Dictionary; 1993
2. W Roger Breed; The Weights and Measures Act, 1963; Charles Knight; 1964
3. The Chambers Dictionary; 1994
4. Various correspondents on the US Metric Association mailing list
5. Various physics textbooks
6. Various history textbooks
7. Wikipedia

Home + About me + E-mail me